Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: DamonJones on November 12, 2011, 02:40:33 PM
-
So I have an old Lagun mill that I've converted to CNC. It works great but I am considering using an AC servo for the spindle. The current setup uses a 5hp 3phase motor at 1750 rpm. It has a variable belt clutch setup that is easy for speed changes but still a manual process. Plus I'd like to get rid of the 3 Phase if possible. So here's the question. What would you consider as a minimum HP or KW rating for the new motor? My thought would be if I pickup max RPM to say 4000 I could lose some HP and keep the cost down. I'd eliminate the adjustable clutch and drive the 2 speed gear box directly. That way I still could have good slow speed torque if needed.
This is a home system so production is not an issue. I don't mind cutting slower but I don't want to cripple the machine either.
-
On my Beaver Mill I put an Allen Bradley MPL-B580J motor which is 8KW 34Nm cont 87Nm Peak. It is well over powered and I dont have any need for the low range in the gearbox, I can put an 8 inch collet spanner on the collet holder and start the spindle at 1RPM and only just manage to stall it, any faster and no chance.
On the Bridgeport, which I am re-doing at the moment, I have put a F4050 Allen Bradley motor which is 6.7Nm cont torque and about 2.8Kw and although I have not really done much with it yet it seems plenty.
Hood
-
What size driver does the F4050 require?
-
DSD-030 although you could probably get away with a DSD-020 if you also had an active shunt.
Hood
-
Could you explain the active shunt to me? Save me a bunch of research time. Thanks
-
The Allen Bradley micro drives do not have internal shunt circuitry so if you try and accelerate too fast or slow down too fast with biggish motors they will trip on Bus overvoltage. The active shunt is a unit that contains the circuitry and also a shunt resistor. I use them with 020 drives on the X and Z of the Beaver mill.
The full size drives have internal circuitry and shunt resistors so an active shunt is not needed and if more shunt resistance is required you can attach to the external terminals.
Hood
-
Here is an active shunt, just so you know what to look for.
http://www.ebay.co.uk/itm/ALLEN-BRADLEY-2090-UCSR-A300-ACTIVE-SHUNT-REGULATOR-/230584345515?pt=LH_DefaultDomain_0&hash=item35afe5e3ab
Hood
-
Is there a particular unit you recommend? Or a rating that is required?
-
Not sure what you are meaning, if its the active shunt you are talking about there is only 1 size as far as I am aware.
Hood
-
Yes the shunt, I'll see what I can find. Sorry AC servos are new to me.
On another note I noticed in one of your other posts talking about line drivers for the DSD drives. What levels are they looking for on STEP and DIR? Do you have a sample schematic of your solution?
-
No schematic as all thats in it is two quad driver chips for 4 axis worth of differential step and direction outputs. The line drivers I use are SN75172.
Hood
-
Perfect I can figure it out from there. I knew that degree in Electronics would come in handy some day. ;D
-
Ha ha well you should manage just fine :D
Hood
-
So I went ahead and picked up the DSD-020-SE, Active shunt and a 4050 motor. Got good deals on them all and hope they all work out. The project may take a while however as I need to convert to a Smoothstepper from my current PP setup. Been wanting to do that anyway and will need it now for sure. I've blown my budget for a while here however so that will have to wait. If anyone has one they want to let go cheap I'm all ears. Should give me time to figure out mounting, gearing and the software for the driver anyway. Thanks again for the help.
-
You can still use the parallel port until you can afford a SS, in fact it may be wise to wait as there is a new version of the SS that should be out soon, it will be an Ethernet version.
Ok back to using the parallel port, you can use electronic gearing in the drives, that way one pulse from Mach is multiplied by the amount you have set and this will get you the Velocities you require. The drawback is the resolution is decreased and if set too high the motion may be coggy at very low revs. Its not something I like to do but I did use it on the lathe at 2:1 before the SS came out.
Hood
-
Funny you bring it up. I decided last night that I'm going to play with the PP first to get the most performance I can out of it and to try the gearing you mentioned. I happened across that feature while reading the drive manual. It will at least let me get the project up and going sooner.
I like the idea of an Ethernet based controller there seems to be a big push for that in the market that Smoothstepper caters to. Does anyone know the status of the project?
-
Funny you bring it up. I decided last night that I'm going to play with the PP first to get the most performance I can out of it and to try the gearing you mentioned. I happened across that feature while reading the drive manual. It will at least let me get the project up and going sooner.
I like the idea of an Ethernet based controller there seems to be a big push for that in the market that Smoothstepper caters to. Does anyone know the status of the project?
I know, but I'm not allowed to say.... :-)
Regards,
Ray L.
-
Funny you bring it up. I decided last night that I'm going to play with the PP first to get the most performance I can out of it and to try the gearing you mentioned. I happened across that feature while reading the drive manual. It will at least let me get the project up and going sooner.
I like the idea of an Ethernet based controller there seems to be a big push for that in the market that Smoothstepper caters to. Does anyone know the status of the project?
I know, but I'm not allowed to say.... :-)
Regards,
Ray L.
Ahh come on now no secrets. :)
-
So I just went out and played with my clocking on the mill. The P4 3GHZ motherboard will only give me 45Khz without freezing during start up. I could upgrade it but that seems like more money and work then going to the SS and being done with the issue.
For now I'll use what I have to test the setup. Assuming all the Ebay second hand stuff works then I'll invest in the SS. I also have to fab in the new motor, gears and belts, ect. Just more money that will have to wait on for now.
So does anyone want to figure out what RPM I can get on a 4050 at a 45Khz pulse rate?
-
The Kernel wont limit the RPM if you use electronic gearing in the drive and with it being a spindle electronic gearing wont really affect things as it would an axis as you will be unlikely to be rotating at very low speeds where cogging would be evident.
You may however not get quite the max RPM out of the motor with it being on an 020 drive so shoot for 3500 rpm and hope for the full 4000 :)
4000rpm x 8000 pulses per rev = 32,000,000 pulses per min divided by 60 = 5,333,333 pulses per second
5,333,333 / 45,000 = 11.85 so a gearing of 1:12 in the drive would allow you full speed if you can get it.
Hood
-
Damon,
Just got my test Ethernet SS today :)
Looking good so far although I have only tested in simulation as I will need to do a wee bit of rewiring for my mill.
Just talked to Greg and he says probably beginning to mid Dec when they will be ready but that is dependant on some parts he is waiting for arriving on time.
Hood
-
That way cool - Perfect timing as I should have funds by then.
Hey off the top of your head. Does swapaxis also move tuning params or just the ports? This will become an issue when it comes time to do my 4th axis project as I plan to turn on it as well. I thought I read somewhere that all it does is move the ports. This will get a bit tricky setting up if so. Especially if they end up geared differently. Haven't thought it all the way through yet but I'm sure it will be an issue. Of course there will be a bunch of factors depending on the drives, motors and encoders that I end up with.
-
SwapAxis just swaps the pin numbers around.
Hood
-
Ya that's what I thought. Defiantly going to want the SS when the time comes. No way the 1.85 steps per unit with the 12:1 gearing is going to work on a 4th axis as spindle. I'd imaging having it geared down some where between 2:1 and 4:1 making the slow 45K even that much worse. Although turning slower would be fine I guess but the RPM's in Mach will be off. That's going to be a problem anyway unless I can balance the two systems. That's going to take some thought. I'd imagine setting things up to work as desired on the 4th axis and then using the DSD gearing to balance the two. Guess I'll work that out when the time comes. To much to think about for now.
-
Not really sure what you are wanting with the 4th axis but unless motor tuning is the same (accel, Vel Steps per) as the axis you are to swap with then I think you would have issues.
Hood
-
My thought exactly however I was looking over the Wiki last night and noticed the motor tuning params can be changed via macro. So if that does become an issue I should be able to changed those in my macros that swapaxis. As far as gearing differences go I figure I can use different pulley setups to compensate. Also changeable via Macro by changing the DRO.
-
That should work well :)
Hood